World Anti-Doping Agency (WADA) v. Gil Roberts

Arbitration CAS 2017/A/5296 World Anti-Doping Agency (WADA) v. Gil Roberts, award of 25 January 2018

Panel: The Hon. Hugh Fraser (Canada), President; The Hon. Michael Beloff QC (United Kingdom); Mr Jeffrey Benz (USA)

Athletics (sprint) Doping (probenecid) Meaning of the balance of probability test No fault or negligence

1. In case a panel is offered several alternative explanations for the ingestion of a prohibited substance, but is satisfied that one of them is more likely than not to have occurred, the athlete has met the required standard of proof regarding the means of ingestion of the prohibited substance. It remains irrelevant that there may also be other possibilities of ingestion, as long as they are considered by the panel to be less likely to have occurred. In other words, for the panel to be satisfied that a means of ingestion is demonstrated on a balance of probability simply means, in percentage terms, that it is satisfied that there is a 51% chance of it having occurred. The athlete thus only needs to show that one specific way of ingestion is marginally more likely than not to have occurred.

2. A panel cannot find that the athlete did not exercise utmost caution when he kissed his girlfriend of three years after she had, without his knowledge, consumed medication containing the probenecid ultimately found to be present in his system. The athlete could not have envisioned that she might be inadvertently responsible for administering the probenecid to him if he were to kiss her prior to his out-of-competition test. It follows that the athlete acted without fault or negligence.

I. PARTIES

1. The World Anti-Doping Agency (“WADA” or the “Appellant”) is a Swiss private law foundation with its headquarters in Montreal, Canada, and its seat in Lausanne, Switzerland, whose object is to promote and coordinate the fight against doping in sport in all its forms.

2. Gil Roberts (the “Athlete” or “Respondent”) is an American athlete specializing in the 200m and 400m sprint events, and a member of the gold medal 4x400m relay team during the 2016 Rio Olympic Games.

II. FACTUAL BACKGROUND — A. Background Facts

3. Below is a summary of the relevant facts and allegations based on the parties’ written submissions, pleadings and evidence adduced at the hearing. Additional facts and allegations found in the parties’ written submissions, pleadings and evidence may be set out, where relevant, in connection with the legal discussion that follows. While the Panel has considered all the facts, allegations, legal arguments and evidence submitted by the parties in the present proceedings, it refers in its Award only to the submissions and evidence it considers necessary to explain its reasoning.

4. On 24 March 2017, the United States Anti-Doping Agency (“USADA”) collected an out-of- competition urine sample from the Athlete. The Athlete’s sample was subsequently sent for analysis to the WADA-accredited laboratory in Los Angeles, California.

5. The laboratory later reported that the Athlete’s A Sample had tested positive for probenecid with a concentration estimated at 9ng/mL. Probenecid is a Specified Substance in the class of Diuretics and Masking Agents on the WADA List of Prohibited Substances.

6. On 14 April 2017, USADA notified the Athlete of the results of his A Sample. The letter gave the Athlete the option to accept a provisional suspension.

7. On the same day, 14 April 2017, USADA requested that the Athlete provide an explanation as to how probenecid entered his body.

8. On 19 April 2017, the Athlete’s agent informed USADA that the Athlete had withdrawn from two upcoming events, namely the IAAF World Relays and the Penn Relays. The Athlete signed an official provisional suspension form on 5 May 2017.

9. On 1 May 2017, the Athlete was informed that his B Sample confirmed the results of his A Sample.

10. Also on 1 May 2017, USADA referred this matter to the Anti-Doping Review Board to determine whether there was sufficient evidence to charge the Athlete with an anti -doping rule violation (“ADRV”). The Athlete was thereafter invited to file a written submission in support of his defense.

11. On 27 May 2017, the Athlete retained Mr. Paul Greene as his legal counsel in this proceeding.

12. On 1 June 2017, the Athlete sent a package containing one single yellow capsule to the Banned Substances Control Group (“BSCG”) Laboratory in Los Angeles for analysis.

13. On 7 June 2017, the Athlete filed his written submission which included test results from the BSCG Laboratory indicating that a “Moxylong” capsule had been tested which contained probenecid. The submission also included witness statements indicating that the Athlete believed the positive test results for probenecid resulted from kissing his girlfriend who had

recently ingested Moxylong capsules sublingually immediately before he kissed her and shortly before his out-of-competition doping control.

B. The Ingestion of Probenecid

14. The Athlete asserts that from the period 7 – 17 March 2017, his girlfriend, Ms. Rebecca “Alex” Salazar, was on holiday with her family in India when she sustained a sinus infection. As a result, Ms. Salazar, with the assistance of her step-father, Mr. Rizwan Siddiqi, visited a pharmacy in “semi-rural India” to obtain medication. Mr. Siddiqi, who speaks Hindi, explained his daughter’s condition to the pharmacist and requested a form of medicine in a capsule so she could pour the medicine directly in her mouth (as opposed to a pill which would require swallowing, which she had difficulty doing), which was her preferred method of ingesting medications.

15. The product provided by the pharmacist was called Moxylong. Ms. Salazar was aware the product contained an antibiotic but was not aware that it also contained probenecid.

16. Ms. Salazar took the Moxylong daily from 14-28 March 2017 (i.e. including for several days beyond the Athlete’s doping control). On the day of the doping control, Ms. Salazar asserts that she ingested her daily dose of Moxylong shortly after arriving at the Athlete’s residence sometime between 13h00 and 13h30. Her ingestion followed her normal procedure whereby she manually opened the capsule and poured the contents onto her tongue and washed it down with water.

17. From 13h00 onward the Athlete asserts that he engaged in “a lot” of kissing with Ms. Salazar, including momentary periods of time after the arrival of the doping control officer and just before undergoing the doping control test.

18. The Athlete did not know that Ms. Salazar was taking Moxylong and did not see her take any of her medication.

19. Following notification of the Athlete’s ADRV, and the Athlete’s consequent efforts to determine the source of the probenecid, Ms. Salazar retrieved her medicine kit in her travel bag where she kept her Moxylong capsules. The original packaging for the Moxylong had been discarded. Ms. Salazar had, however, one capsule remaining, which was sent by the Athlete to the BSCG Laboratory for testing.

C. Proceedings before the American Arbitration Association

20 On 12 June 2017, USADA sent the Athlete a charge letter and an expedited hearing ensued based upon the agreement of the parties.

21. On 20 June 2017, a hearing was held by telephone conference before Hon. John Charles Thomas with the American Arbitration Association (“AAA”) in accordance with the USADA Protocol for Olympic and Paralympic Testing, effective 1 January 2014 (“USADA Protocol”). The USADA Protocol implements the World Anti-Doping Code for US athletes. The Athlete

testified at the hearing as did his girlfriend. Dr. Pascal Kintz testified as an expert witness for the Athlete. Dr. Matthew Fedoruk testified as USADA’s expert witness.

22. Later that same day, Judge Thomas issued an “Operational Award” determining that the probenecid in the Athlete’s system was a result of kissing his girlfriend and that the Athlete had no way of knowing that he was exposing himself to a doping violation in so doing. As a result, a finding of No Fault or Negligence was made.

23. A reasoned award was later issued on 10 July 2017 (the “Appealed Decision”).

III. PROCEEDINGS BEFORE THE COURT OF A RBITRATION FOR SPORT

24. On 24 August 2017, the Appellant filed its statement of appeal against USADA and the Athlete with respect to the Appealed Decision in accordance with Article R47 et seq. of the Code of Sports-related Arbitration (the “Code”). In its statement of appeal, the Appellant nominated the Hon. Michael J. Beloff M.A. Q.C. as arbitrator.

25. On 1 September 2017, the Respondent nominated Mr. Jeffrey G. Benz as arbitrator.

26. On 5 September 2017, the Appellant, upon agreement of the parties, agreed to withdraw its claims against USADA

27. On 4 October 2017, the CAS Court Office, on behalf of the President of the Appeals Arbitration Division, confirmed the appointment of the Panel as follows:

President: Hon. Hugh L. Fraser, Judge, Ottawa, Canada Arbitrators: Hon. Michael J. Beloff M.A. Q.C, London, United Kingdom Mr. Jeffrey G. Benz, Attorney-at-Law, Los Angeles, California and London, United Kingdom.

28. On 9 October 2017, following an agreed-upon extension of time between the parties, the Appellant filed its appeal brief in accordance with Article R51 of the Code.

29. On 11 December 2017, the Respondent filed his answer in accordance with Article R55 of the Code.

30. On 9 and 10 January 2018, an Order of Procedure was signed by the Respondent and Appellant, respectively.

31. On 15 January 2018, a hearing was held at the offices of JAMS, New York Times Building, 620 Eighth Avenue, New York, New York. The Panel was assisted at the hearing by Mr. Brent J. Nowicki, Managing Counsel to CAS. The following persons were in attendance:

For the Appellant: Mr. Ross Wenzel, Counsel for WADA Mr. Adam Klevinas, WADA Senior Legal Manager Dr. Olivier Rabin, WADA Senior Science Director Dr. Matthew Fedoruk, USADA Senior Managing Director of Science and Research Dr. Alka Beotra (expert witness)

For the Respondent: Mr. Gil Roberts, Appellant Mr. Paul Greene, Counsel for Mr. Roberts Mr. Matthew Kaiser, Counsel for Mr. Roberts Ms. Juhi Gupta, legal intern Ms. Rebecca “Alex” Salazar, witness Mr. Rizwan Siddiqi, witness (by telephone) Dr. Pascal Kintz (expert witness) Dr. Anand L. Kulkarni (expert witness) (by Skype).

32. At the outset of the hearing, the parties confirmed that they had no objection to the composition of the Panel. During the hearing, the Panel heard evidence from the witnesses listed above including the five experts in addition to the detailed submissions of counsel.

33. At the conclusion of the hearing, the Parties indicated that they were satisfied that their right to be heard had been duly respected and they had been treated fairly and equally during the arbitration proceedings.

IV. SUBMISSIONS OF THE PARTIES — A. The Appellant

34. The Appellant’s submissions, in essence, may be summarized as follows:

- There is no dispute between the parties that the Athlete committed an ADRV when he tested positive for probenecid. With respect to such ADRV for this specified substance, a four-year period of ineligibility may be imposed in the event the Anti-Doping Organization establishes that the ADRV was intentional. In this appeal, the Appellant concedes that it cannot demonstrate that the ADRV was intentional and therefore, it does not request a four-year period of ineligibility.

- Instead, the Appellant asserts that the default two-year period of ineligibility applies to the Athlete’s ADRV in accordance with Article 10.2.2 of Annex A to the USADA Protocol. Such default period of ineligibility may be eliminated or reduced in the event that the Athlete establishes that he bears No Fault or Negligence (Article 10.4) or that he

fulfills the conditions of either Article 10.5.1.1 (Specified Substances) or Article 10.5.2 (No Significant Fault). In order to qualify for the application of these provisions, the Athlete must demonstrate how the prohibited substance (probenecid) entered his body on the balance of probability.

- The Appellant submits that the Athlete’s explanation for the presence of probenecid in his system is replete with inconsistencies and implausible explanations. It maintains that there are many properly run professional pharmacies in India where Ms. Salazar could have obtained medication other than Moxylong to treat her sinus infection. They also point to the evidence indicating that Moxylong has not been manufactured since 2003 and question accordingly whether an off-road pharmacy in a small town would have this medication in stock, ready to dispense to Ms. Salazar.

- The Appellant also submits that one of their expert witnesses, Dr. Beotra, carried out her own extensive search for Moxylong during a two-week visit to India and was unable to locate any pharmacy that carried this medication or that was even familiar with Moxylong. The Appellant adds that Dr. Beotra also made several attempts to purchase Moxylong on-line and was unsuccessful.

- The Appellant submits that if Ms. Salazar’s step-father, Mr. Siddiqi was as concerned about his stepdaughter as he professed to be, he would have found a more reputable location to purchase an antibiotic for Ms. Salazar or at the very least would have asked many more questions about what was being given to her to treat her sinus condition.

- The Appellant further submits that someone presenting with Ms. Salazar’s symptoms would have many other options available including common amoxycillin products that would have the same or better therapeutic results than Moxylong. It is also skeptical of the Respondent’s witnesses’ testimony that the Moxylong tablet was ground into powder and placed into capsules by the pharmacist in India to make it easier for Ms. Salazar to ingest. The Appellant underscores the fact that Moxylong dissolves in water and would not need to be ground into powder.

- The Appellant finds it unlikely that Ms. Salazar, who professed a dislike of antibiotics, would nonetheless have continued to take the capsules after she felt better, but then adventitiously retained only one.

- The Appellant also submits that the science behind the Respondent’s evidence is unreliable. They point to what they describe as a massive disparity between the quantity of probenecid found in his system and the quantity of probenecid that was found in the capsule analyzed by BSCG.

- The Appellant’s experts maintain that the normal process of salivating reduced the amount of the probenecid that would be retained in one’s mouth dramatically so that if Ms. Salazar took a drink of water after ingesting the capsule, that along with the normal process of salivating would leave a negligible amount of probenecid to be transferred by kissing after a few minutes.

- The Appellant questions whether any capsule taken by Ms. Salazar was even Moxylong based on the fact that Moxylong was comprised in equal parts of amoxicillin and probenecid (e.g. 500 mg of each substance for the standard size pills). The Appellant notes that the pill analyzed by BSCG laboratory, said to be the last of the course of Moxylong, contained 8.7 micrograms probenecid out of a total weight of 442,000 micrograms and concludes that the capsule would have had to contain almost 57 thousand times more probenecid in order to be comprised of 50% of that substance.

- Lastly, the Appellant submits that the evidence given by the Respondent’s expert witness, Dr. Pascal Kintz, has no scientific basis and was altered to meet the changing facts in the Athlete’s explanation.

35. In their appeal brief, the Appellant requested the following relief:

WADA hereby respectfully requests CAS to rule that: 1. The Appeal of WADA is admissible.

2. The decision of the American Arbitration Association dated 10 July 2017 in the matter of Gil Roberts is set aside.

3. Gil Roberts is sanctioned with a two-year period of ineligibility starting on the date on which the CAS award enters into force. Any period of provisional suspension effectively served by Gil Roberts before the entry into force of the CAS award shall be credited against the total period of ineligibility to be served.

4. All competitive results obtained by Gil Roberts from and including 24 March 2017 until 20 June 2017 are disqualified, with all resulting consequences (including forfeiture of medals, points, and prizes).

5. The costs of the arbitration are borne by the Respondent.

6. WADA is granted a significant contribution to its legal and other costs.

B. The Respondent

36. The Respondent’s submissions, in essence, may be summarized as follows

- As amplified below, the Athlete has met the required standard of proof regarding the means of ingestion of probenecid by providing evidence that kissing Ms. Salazar was the likely source of contamination/ingestion.

- He tested positive for an extremely low amount of probenecid (less than 10 ng/ml) that “would be useless for cheating”. Dr. Kintz opines that the level in the Athlete’s urine is consistent with incidental exposure to minute amounts of probenecid a few hours prior to his urine test on the afternoon of 24 March 2017.

- Approximately 3 hours prior to the Respondent’s positive test on 24 March 2017, Ms. Salazar took a medication that contained probenecid, called Moxylong, which she had obtained at a pharmacy in India less than two weeks prior to the Respondent’s doping control test. Ms. Salazar took the Moxylong (as she takes all medicines) by opening the capsule onto her tongue and then washing it down with the water bottle she was sharing with the athlete.

- Mr. Roberts and Ms. Salazar were intimate and kissed for much of the afternoon leading up to his drug test and almost certainly kissed within a few minutes of when she took the Moxylong containing probenecid.

- Mr. Roberts drank from the same water bottle as Ms. Salazar (which she used to wash down the Moxylong containing probenecid) throughout the afternoon prior to his drug test.

- A laboratory test performed by BSCG on Ms. Salazar’s remaining Moxylong capsule established the presence of probenecid in the medicine she took.

- Mr. Roberts tested all the supplements he was taking and none contained probenecid.

- There is no other possible incidental source for the trace amount of probenecid that Mr. Robert had in his 24 March 2017 urine sample other than Ms. Salazar.

- WADA has concluded that this is not a case of intentional use.

- Dr. Kintz concluded in his second report that the statement of Dr. Rabin had not considered the athlete’s physiology, metabolism and drug distribution. Dr. Kintz again confirmed that the most likely source of the probenecid is a contaminated kiss with Ms. Salazar on the afternoon of 24 March 2017.

- Dr. Kulkarni concluded that it was entirely possible for Ms. Salazar in March 2017 to be given Moxylong, without a prescription, in capsule form that did not directly correspond to the tablet form in order to treat a sinus infection, just as Ms. Salazar recounts.

- The Respondent further submits that even with the exercise of utmost caution, he could never have known that kissing Ms. Salazar could lead to a positive test. He adds that he was in no position to know that it was medically possible to be contaminated with probenecid by kissing someone who had ingested it.

- The Respondent submits that it is uncontested that one can obtain antibiotics without a prescription in small pharmacies in India and that Dr. Beotra’s evidence that one cannot find empty capsules in all of India is absurd, unreasonable and contradicted by Dr. Kulkarni’s personal experience.

- The Respondent also asserts that the Panel should prefer the evidence of Dr. Kulkarni who has been a licensed physician in India since 1981 to that of Dr Beotra, who is not medically qualified to assert such a statement.

- The Respondent states that it is impossible to know the level of what was found in the Moxylong capsule because there was no fair quantification of its contents. He acknowledges that Dr. Rabin and Dr. Fedoruk testified in a fair manner but notes that the Appellant’s own witnesses recognized that the BSCG laboratory did not follow Standard Operating Procedure, with the consequence that no accurate quantification of the capsule was performed.

- The Respondent maintains that neither he nor Ms. Salazar would have had enough time to make up such a story given the short turnaround between the time when he retained counsel (over the Memorial Day holiday weekend in the United States) and the shipping of the remaining Moxylong tablet to the BSCG laboratory on 1 June 2017.

- The Respondent further submits that the fact that WADA has not tried to prove intentional ingestion is significant as is the fact that probenecid on its own is virtually useless to an athlete and its use is only recorded in unintentional use cases.

- Lastly, the Respondent submits that he has proven the origin of the prohibited substance on a balance of probability and should therefore be found to bear no fault or negligence.

37. In its answer, the Respondent requested the following relief:

WHEREFORE, Mr. Roberts respectfully requests that this CAS Panel (1) Finds he bears No Fault or Negligence and eliminate his period of ineligibility so that he is immediately eligible to compete;

(2) In the alternative, reduce his period of ineligibility based on No Significant Fault or Negligence and sanction him with a reprimand and no period of ineligibility so that he is immediately eligible to compete;

(3) In the event Mr. Roberts is sanctioned with 24 [month] period of ineligibility, give Mr. Roberts credit for the voluntary provisional suspension he has served pursuant to Article 10.11.3 and render him eligible on 16 July 2019,

(4) In the alternative, determine that Mr. Roberts should commence his period of ineligibility on 24 March 2017, the date of sample collection, pursuant to Article 10.11.2 and render him eligible on 4 June 2019;

(5) Preserve all medal, points, and prizes earned by Mr. Roberts in the period between the issuance of the operative award on 20 June 2017 and the renewal of his voluntary provisional suspension on 1 September 2017,

(6) Order any other relief for Mr. Roberts that this Panel deems to be just and equitable including an award of fees and costs.

V. JURISDICTION

38. Article R47 of the Code provides as follows:

An appeal against the decision of a federation, association or sports-related body may be filed with the CAS insofar as the statutes or regulations of the said body so provide or as the parties have concluded a specific arbitration agreement and insofar as the Appellant has exhausted the legal remedies available to him prior to the appeal, in accordance with the statutes or regulations of the said sports-related body.

39. The Appellant relies on Article 17 B) of the USADA Protocol (as set forth in Article 13.2 of Annex A thereto), as well as R-45 of the AAA Supplementary Procedures (also attached as Annex D to the USADA Protocol) as conferring jurisdiction on the CAS. Moreover, the Appellant refers to Article 13.2.1 of the WADC as a further means of granting jurisdiction to the Panel to decide this appeal.

40. The Respondent does not dispute jurisdiction and indeed, confirmed CAS jurisdiction in signing the Order of Procedure and participated in this proceeding fully.

41. Consequently, the Panel determines that it has jurisdiction to decide this appeal.

VI. A DMISSIBILITY

42. Article R49 of the Code provides as follows:

In the absence of a time limit set in the statutes or regulations of the federation, association or sports-related body concerned, or of a previous agreement, the time limit for appeal shall be twenty-one days from the receipt of the decision appealed against. After having consulted the parties, the Division President may refuse to entertain an appeal if it is manifestly late.

43. Pursuant to Article 17 b) of the USADA Protocol, “subject to the filing deadline for an appeal filed by WADA as provided in Article 13.2.3 of the Code, the final award by the AAA Arbitrator(s) may be appealed to CAS within twenty-one (21) days of issuance of the final reasoned award …”.

44. According to Article 13.2.3 of the WADC, “the filing deadline for an appeal filed by WADA shall be the later of (a) Twenty-one days after the last day on which any other party in the case could have appealed; or (b) Twenty-one days after WADA’s receipt of the complete case file relating to the decision”.

45. The International Association of Athletics Federations, which has a right to appeal under Article 13.2.3 of the WADC, received the Appealed Decision on 14 July 2017. The IAAF’s deadline to appeal, therefore, expired on 4 August 2017. WADA filed its statement of appeal on 24 August 2017.

46. No party disputed admissibility of this appeal and the parties participated in this proceeding fully.

47. Consequently, the Panel determines that this appeal is timely and admissible.

VII. A PPLICABLE LAW

48. Article R58 of the Code provides as follows:

The Panel shall decide the dispute according to the applicable regulations and the rules of law chosen b y the parties or, in the absence of such a choice, according to the law of the country in which the federation, association or sports-related body which has issued the challenged decision is domiciled or according to the rules of law, the application of which the Panel deems appropriate. In the latter case, the Panel shall give reasons for its decision.

49. As set forth above, the Athlete’s doping control procedure was initiated and directed by USADA in accordance with the USADA Protocol. In his doping control form, the Athlete agreed to submit, if needed, to the USADA Protocol. The Athlete has not objected to the contrary in this appeal. Therefore, the Panel determines that this appeal shall be decided on the basis of the USADA Protocol, which is based on the 2015 WADC, which seeks to harmonize anti-doping policies, rules and regulations in all sports in the United States and, at least with respect to the 2015 WADC, globally.

VIII. MERITS

50. The Panel’s starting point is with Article. 2.1 of Annex A of the USADA Protocol, which provides that the presence of probenecid in the Athlete’s sample constitutes an anti-doping rule violation. In this case, the analysis of the Athlete’s A and B samples revealed the presence of probenecid.

51. It is common ground between the parties that, as the Panel accepts, (i) the default period of ineligibility of two years for an anti-doping rule violation involving a specified substance such as probenecid can be eliminated if the athlete establishes that he bore no fault or negligence (Article 10.4); (ii) the athlete must for that purpose establish the origin of the substance (Definitions); and (iii) he must do so on the balance of probabilities (Article 3.1).

52. The Panel adopts the guidance given in CAS 2009/A/1926 & 1930 that:

“… [I]t is the Panel’s understanding that, in case it is offered several alternative explanations for the ingestion of the prohibited substance, but it is satisfied that one of them is more likely than not to have occurred, the Player has met the required standard of proof regarding the means of ingestion of the prohibited substance. In that case, it remains irrelevant that there may also be other possibilities of ingestion, as long as they are considered by the Panel to be less likely to have occurred. In other words, for the Panel to be satisfied that a means of ingestion is demonstrated on a balance of probability simply means, in percentage terms, that it is satisfied that there is a 51% chance of it having occurred. The Player thus only needs to show that one specific way of ingestion is marginally more likely than not to have occurred”.

53. There are two main points for the Panel to consider: (1) Does the Panel accept as more likely than not the explanation advanced on the Respondent’s behalf (the “Roberts version”) that the residue of a course of Moxylong, purchased for his girlfriend in India and consumed by her on the day of his out-of-competition test was transferred to him by kissing; and (2) if so, does the

Panel accept that this was the reason why his urine sample taken on 24 March 2017 contained probenecid at an estimated concentration of 9ng/ml.

54. The Panel notes that it is, in theory, possible for it to accept the first point, but reject the second. This would, however, require it to conclude that there were two sources of probenecid in the Athlete’s system on the same day, a far-fetched coincidence which the Panel, in the absence of any supporting evidence, must discard. The Panel does, however, note that the Athlete must establish the origin of the concentration of probenecid in his system so that what might otherwise be an apparently plausible explanation of origin could be fatally undermined by scientific, in particular pharmacokinetic, evidence. The first and second points are therefore intertwined, not independent.

55. The Panel appreciates that, although WADA have not sought to advance a case that the athlete was guilty of an intentional anti-doping rule violation which would have required evidence to a standard of comfortable satisfaction, it is an important plank of WADA’s case that the Athlete and his lay witnesses have concocted a false story to explain an adverse analytical finding. Moreover, the inference could properly be drawn on that premise that the only logical reason for them to lie would be because, in fact, the probenecid had been taken by the Athlete as a masking agent to conceal the deliberate use of a performance enhancing substance.

56. While the critical question for the Panel is simply whether the Athlete has or has not sat isfied the test put forward in CAS 2009/A/1926 & 1930, the Panel should not and does not ignore the potential implications of a finding one way or the other.

57. The Panel has not found the critical question easy to resolve. Powerful arguments have been advanced by the experienced advocates on both sides; and there is no single undisputed or indisputable fact which tilts the scales in one direction rather than another. A plausible narrative can be constructed in favour of or against WADA’s appeal without either scenario ignoring the precepts of physics or indeed of psychology. To take but one example, it is on its face peculiar that Ms. Salazar should retain one, but only one, capsule of Moxylong out of a prescribed course of antibiotics; but it is not impossible that she should do so. She has provided an explanation which is not palpably absurd but which demands evaluation in the overall evidential context. This is a case in which the law may be clear, but the facts less so.

58. Reverting to the first point, the Panel notes that certain of the necessary (if not sufficient) elements of the Athlete’s version are not in issue:

(i) Ms. Salazar is the Athlete’s girlfriend and they have been in a relationship for the past three years;

(ii) Ms. Salazar, with her step-father and mother, did take a week-long vacation in India;

(iii) Ms. Salazar’s step-father is of Indian ethnic origin and is a frequent visitor to India;

(iv) Indian pharmacies are of different quality and many of those not in major city areas are of unimpressive appearance and organization;

(v) In pharmacies of that latter kind, antibiotics can be obtained without prescription (whereas prescriptions would be required in better quality pharmacies);

(vi) Moxylong was manufactured in India until 2003, and there was no legal requirement to cease its sale thereafter; indeed it is still advertised on websites (albeit the Panel was not provided with evidence from either side that it had been purchased other than on the occasion alleged by the Athlete). Indeed, Moxylong is still listed on the official Indian Monthly Index of Medical Specialities (MIMS);

(vii) Moxylong contains probenecid; and

(viii) Ms. Salazar has a history of respiratory and associated symptoms. The Panel had no evidence to this effect other than from her and her family but that evidence was not challenged by WADA.

59. The Panel repeats that none of these elements makes the Athlete’s version of the story by itself even marginally more probable than a different explanation for the adverse analytical finding. Other and more detailed evidence was required for that purpose, e.g. materially that Ms. Salazar suffered from respiratory problems from the start of her holiday; that she only decided she needed medicine in anticipation of her return trip by air; that the Moxylong was purchased and produced in the circumstances described by her and her step-father; that it was consumed in the manner, over the period and at the times described by her; that a single capsule was retained, and was the one sent to BSCG; that on the day of the test she and Mr. Roberts had enjoyed intimate relations accompanied with repeated kissing.

60. On consideration of the totality of the evidence, the Panel is disposed to accept the Athlete’s evidence on these matters for the following reasons:

(i) To reject it requires the Panel to find in fact if not in strict law, that both the Athlete and more importantly, the other lay witnesses, conspired to mislead the Panel. This would include Ms. Salazar’s mother, who, though she was not available to provide evidence at the hearing, nonetheless gave a written statement consistent with that of her daughter and husband, which WADA expressly did not accept to represent the whole truth. Though the stakes for the Athlete were high, the Panel would be presumptively reluctant to find that there was such a conspiracy, given the gravity of such finding, though obviously open to so doing if that was indeed their informed view of the witnesses’ testimony.

(ii) The Panel had the opportunity to evaluate that testimony and the demeanor of the main witnesses, Ms. Salazar and the Athlete, by sight and sound. While the Panel is aware that the least honest witnesses can appear the most plausible (and vice versa), the Panel judged them to be truthful, not least because they did not retreat from their evidence in any material way under sustained and powerful cross-examination. There were no telltale traces of a rehearsed version of their evidence, which was indeed given somewhat artlessly, nor anything in their background which would cast doubt on their believability.

(iii) Appellant’s counsel initially drew attention to the unconventional, extra -sensory belief system of Mr. Siddiqi, but this did not in the Panel’s view, bear upon his credibility, vis- a-vis his evidence as to what may have happened on the road between New Delhi and Jaipur and Mr. Wenzel did not press the point. The Panel did, however, note that Ms. Salazar, who had described her step-father as a film director, appeared wholly unaware of his actual occupation, which would be odd if they had pre-rehearsed an untruthful story.

(iv) If the Athlete was aware that he had taken some product which was or which conta ined a prohibited substance, the tests which he commissioned from BSCG on a package received by them on 2 May 2017 to ascertain whether any of the supplements he regularly consumed could themselves have contained or been contaminated by prohibited substances would have been to his knowledge, a wholly vain exercise and waste of time and indeed money. No benefit could have accrued to him from the commissioning of such tests, unless they identified the cause of the adverse analytical finding. On the contrary, such behavior would only have been consistent with his genuine ignorance as to what had caused the adverse analytical finding.

(v) Mr. Greene, counsel for the Athlete, who has been involved in a number of contamination cases (including two involving acts of kissing), openly admitted that sources of contamination from, inter alia, contact with third parties were part of his normal “checks” when identifying the source of a prohibited substance in a client’s sample. But this admission cannot of itself provide any basis for a conclusion that the Athlete used advice from his lawyer to as a basis for creating - with the help of his girlfriend and her relatives - a fictitious story of such a contact, rather than a genuine investigation as to whether such causative contact had occurred. Moreover, the Panel could not subscribe to any suggestion, if such were indeed made, that Mr. Greene was somehow involved in directing this storyline. His integrity as a lawyer is unquestioned by this Panel.

(vi) The Appellant submitted that it would have taken no more than a Google search to ascertain that Moxylong contained probenecid and to weave a false narrative around that undisputed fact. In the Panel’s view, that greatly underestimated what would have been necessary to contrive the fiction within the narrow time frame over and after the Memorial Day holiday weekend both in terms of obtaining some capsule containing probenecid (which on WADA’s hypothesis would not have been Moxylong) to send to the BSCG laboratory on 1 June 2017, but also to prepare a coherent script for the several scenes in an invented drama with locations in India as well as California. Such an exercise would moreover have required a degree of sophistication that the Panel could not attribute to the alleged conspirators. It is in the Panel’s view telling that Ms. Salazar appeared unaware that the Athlete’s questions to her about whether and what she had recently consumed (and why indeed he was asking the questions at all) sprang from his lawyer’s advice.

(vii) If the capsule provided for analysis to the BSCG laboratory was not the last of a course of Moxylong purchased for Ms. Salazar in India, its provenance remained unexplained.

Where did that capsule containing probenecid come from? How was it procured and by whom in such a short space of time? These questions, highly relevant to WADA’s challenge, were neither explored nor satisfactorily answered.

(viii) BSCG did not examine or establish that the capsule which it analyzed was or contained an antibiotic. That was not its function. But the laboratory did record at least that the package had been “received intact … with no signs of tampering”. Given that Moxylong was manufactured in tablet form but had been, on the Athlete’s version, crushed and replaced in a capsule by a street pharmacist in semi-rural India, the fact that it may not have matched the composition it enjoyed when distributed originally from the manufacturer caused no substantial concern to the Panel. There was some uncertainty, however, as to why the BSCG analysis recorded as “present” quantities of steroids, stimulants and masking agents. Dr. Kintz interpreted that as indicating their actual presence; the Panel preferred the view of the Appellant’s experts Dr. Rabin and Dr. Fedoruk that it indicated the nature of the tests capable of being carried out by BSCG (testing the equipment if you will), not least because the concept of such an all -purpose prohibited substance was novel and because it would be senseless, indeed ironical, if the Athlete had produced a capsule designed to acquit him, which itself served only to raise yet further questions as to what he was said to have innocently ingested.

(ix) The amount of probenecid found in the Athlete’s out-of-competition sample would have no effect as a useful masking agent. Moreover, no plausible explanation for why probenecid would have been used by the Athlete was put forward by WADA.

(x) While the Panel is conscious that there is a first time for everything, including an anti- doping rule violation, it is entitled to observe that the Athlete’s biological passport was normal at all times and expressly did not raise any concerns or denote any suspicious activity.

61. Mr. Wenzel’s assault on the Athlete’s defense was founded on the proposition that the evidence said to support it was inadequate to carry it over the marginal probability threshold. He suggested the following:

(i) The evidence was bereft with inconsistencies, the hallmark of untruthfulness. Putting aside for a moment the expert evidence of Dr. Kintz, to which the Panel will return, the Panel could identify no inconsistencies of such dimensions as to lead to a rejection of the evidence.

(ii) Mr. Siddiqi’s expression of a sense of responsibility for his step-daughter’s health was at odds with his visit to a lower-quality street pharmacy. But it was only at such a pharmacy, as the Panel has already noted, that antibiotics can be obtained without prescription; and there would be no reason to doubt the safety of a medicine that was properly packaged and named.

(iii) Dr. Beotra, was unable, notwithstanding her efforts, to obtain any Moxylong herself, or notwithstanding her enquiries among professional colleagues and among major

pharmacies even to find proof that it was still available in 2017. But while accepting that her evidence was given carefully and in good faith, the Panel could not eliminate the reasonable possibility that Moxylong might very well still be stocked in a street pharmacy, and whether still efficacious a decade and a half after manufacture or the antibiotic of primacy choice (both of which Dr Beotra - who is not a medical doctor - disputed) could still have been dispensed in such an outlet if not, by contrast, in a major city pharmacy. The Panel refers to the evidence indicating that Moxylong is still listed in MIMS and on certain websites as a saleable product. Nor could the Panel accept that empty capsules were not available for purchase in Indian pharmacies. The Panel accepts Dr. Kulkarni’s testimony to the contrary, which was based on his personal experience as practicing physician in India. As a reputable expert, he had no motive to perjure himself in this forum. Dr. Beotra, on whom WADA relied in this context as well, was again seeking to prove a negative, which is always a notoriously difficult task.

(iv) There was no hard, contemporaneous evidence of the purchase of the Moxylong, e.g. a receipt or a retained package. But in the Panel’s view given the circumstances in which it was alleged that the purchase took place, and the absence of any anticipation that the circumstances of purchase might feature in later proceedings, the absence of such evidence loses the significance it might otherwise have had.

(v) Ms. Salazar had no reason to delay taking a capsule until she was in the Athlete’s apartment at approximately 13h00 while the Athlete was showering after his training session, when she could have taken the capsule in her own apartment earlier in the day. But it was plausible that she, in a hurry to reach her boyfriend’s apartment, could have reasonably waited to take her medicine till then.

(vi). Mr. Wenzel also submitted that the Athlete and Ms. Salazar, by emphasizing that they kissed just prior to the test, showed a lack of understanding of the excretion period of the substance as they attempted to align their evidence. The Panel is unable to attribute to either the Athlete or his girlfriend any purported understanding, correct or incorrect, of such pharmacokinetic science.

62. The Appellant also submitted that the evidence about the crushing of the tablet said to have been purchased, so as to put it into capsule form, was an embellishment of evidence given in the first-instance proceeding. Without a transcript of the earlier proceedings (at which Mr. Wenzel himself was not present) the Panel cannot safely conclude that there was such embellishment, nor indeed does Judge Thomas’s award itself appear to be based on some different version of the purchase.

The Science

63. A powerful argument made by the Appellant was that the Athlete’s story “fails on the science”. In examining the science, the Panel had been asked to consider this essential question: Assuming that Ms. Salazar did indeed at around 13h00-13h30 on 24 March 2017, ingest a capsule of the type analyzed by BSCG laboratory in the manner that she claims (i.e., pouring the contents on her tongue and washing down with water) and then kissing the Athlete over the course of the

next three hours prior to doping control, could she have passed sufficient quantities of probenecid to the Athlete such that he would produce a urinary concentration of 9 ng/ml at 16h16.

64. As stated earlier, the BSCG laboratory performed a qualitative screen test on the package received by them on 2 June 2017. A sample of the item was extracted and analyzed for probenecid according to their in-house methods. The screen testing results indicated the presence of probenecid in the item. No quantity estimate was initially provided since only qualitative screen testing was performed (see BSCG report 6 June 2017). The material provided was used up in the course of the qualitative analysis and nothing remained for further analysis (see BSCG report 14 June 2017).

65. BSCG was subsequently asked to provide a quantity estimate of the probenecid as well as supporting data and chromatograms. Supporting data and chromatograms were provided on 14 June 2017. The author of the analytical report which was provided on 19 June 2017 stated that “in order to provide a quantity estimate at this point we had to compare the qualitative screen testing results with a spiked standard for probenecid run on a different day and do a single point linear estimate based on the standard. Estimating quantity in this way is not in accordance with our normal SOPs and the estimate provided will be a rough estimate only”. The report goes on to say “based on a standard spiked with probenecid at 50 ng/g a rough single point linear estimate of the quantity of probenecid in the item tested is 19,821.938 ng/g”.

66. The quantity estimate report prepared by BSCG was delivered on 19 June 2017, just one day before the AAA hearing. Dr. Kintz who provided expert testimony for the Athlete at that hearing, had prepared a report on 12 June 2017 which appeared to assume that the Athlete would have ingested a maximum of 0.4% of the total amount of probenecid contained within the capsule contained by Ms. Salazar if the expected amount of 250 mg or 500 mg had been discovered in the capsule.

67. The Appellant submits that after discovering that the tested capsule contained only a tiny fraction of the probenecid that he had originally assumed would be present, Dr. Kintz changed his opinion and suggested a range of between 0.4% and 5% ingestion by the Athlete would have been possible.

68. Both Dr. Rabin and Dr. Fedoruk disputed Dr. Kintz’s 5% figure as being unreasonably high and without scientific basis. They further submit that even if 5% of the powder did remain in the oral cavity of Ms. Salazar, common sense dictates that less than the entirety of that amount would be transferred from her to the Athlete as a result of a process in which saliva would move in both directions.

69. The Appellant states that the Athlete’s urine sample was comprised of 165 ml of urine and the amount of probenecid in the urine sample is more than three times the total amount that the Athlete would be expected to have ingested even based on a generous 5% assumption. This assumption becomes all the less plausible, argues the Appellant, when one considers that only a small fraction of the total probenecid dose is excreted in the first three hours post ingestion (with the peak excretion period occurring between 6 and 8 hours post ingestion).

70. Dr. Rabin adds, furthermore, that the capsule analyzed by the BSCG laboratory does not correspond to Moxylong.

71. Dr. Kintz prepared a second report or statement dated 9 December 2017, in response to the report prepared by Dr. Rabin. In Dr. Kintz’s second report, he reaffirmed his conclusion that a contaminated kiss with Ms. Salazar was the most likely source of the probenecid in the Athlete’s 24 March 2017 urine sample.

72. In this second report, Dr. Kintz affirms that the probenecid was correctly ide ntified in the submitted item using two different methods, but that it was incorrectly quantified. Dr. Kintz opines that the quantitative results as provided on 19 June 2017 by BSCG are wholly unreliable because:

• The aliquot was extracted by liquid/liquid extraction. There is no information as to the procedure (pH, solvents), or the percentage of extraction (recovery). This has not been checked, validated and compared with authentic Moxylong preparations.

• During the LC/MS analyses, matrix effects were not investigated. No control was run (and curiously, numerous drugs were identified, such as anastrozole, modafinil, venlafaxine, 19-norandrosterone …).

• There is no data that supports calibration and linearity. There is no data that presents QC specimens. Precision has not been verified. Although indicated in the report of 19 June 2017, there is no data for the single calibration point. There is no peak area comparison with authentic probenecid standard.

73. Dr. Kintz concludes, therefore, that one cannot use the BSCG concentration to evaluate the amount of probenecid present in the capsule and therefore the amount that has contaminated the Athlete.

74. Dr. Kintz concedes that the Athlete was likely not to be exposed to residue from an authentic Moxylong product for two reasons. First, because there is no Moxylong manufactured in capsule form and second, because the measured probenecid concentration is very different from the expected concentration. He considers the possibility that if the capsules were sold under the name Moxylong they should be considered as a counterfeit medicine or a self-made form by a pharmacist.

75. Dr. Kintz noted that the capsules taken by Ms. Salazar would not exactly match the tablets if they were of a “galenic” (or compounded) nature as opposed to authentic tablets. On this point, Dr. Kulkarni also stated “it is not surprising that the weight of the Moxylong capsule Ms. Salazar had remaining (442,000 micrograms) does not directly correspond to a tablet form of Moxylong (either 500,000 micrograms or 1,000,000 micrograms) since the capsule was compounded manually by a pharmacist” .

76. Dr. Kintz added that there is no controlled study to evaluate with precision the amount of probenecid in the oral cavity after pouring the drug on the tongue. The oral cavity is fully contaminated immediately after the drug is poured on the tongue and washes out over time,

with only a small part present after a few minutes. He adds that if there was no kissing for fifteen or twenty minutes there is no substance remaining in the mouth. The contamination would be gone.

77. In Dr. Kintz’s opinion, no one knows the amount of probenecid that will produce a concentration of 9 ng/ml in a urine specimen collected about 3 hours after contamination but the expected dosage range would be between 0.1 to 2 mg, which is an extremely small amount in order to obtain such a low urinary level.

78. The Panel notes that it was the Athlete’s management company who submitted the request to BSCG to test the item described as Moxylong. Mr. Greene agreed that these test results regarding the quantity estimate were not favourable to his client’s position but he felt that he had a duty to disclose those results and he did so. Nevertheless as part of his submissions, he was properly critical of an obvious shortcoming in those quantity estimates.

79. All parties agreed that, on hindsight, the “Moxylong” sample should have been sent to a WADA-accredited lab. Moreover, the quantity estimate conducted by the BSCG laboratory was admittedly not performed in accordance with their normal standard operating procedures (“SOP”).

80. The Appellant submits that although the laboratory did not follow SOP, that does not mean that they operated outside of accepted standards and the results can still be considered reliable. In other words, the Appellant maintains that even accepting Dr. Kintz’s critique the BSCG lab results could not be off to such a degree (a factor of over 25,000) that they would produce the findings contained in their report.

81. The Panel acknowledges that Dr. Rabin and Dr. Fedoruk testified in a very fair manner. Dr. Rabin admitted that when SOP’s are not followed it opens up certain possibilities although he still maintained that the Panel could rely on the BSCG results as reasonably accurate.

82. However, the Panel finds it significant that no accurate quantification of the capsule was performed. To quote again from the letter of 19 June 2017 from BSCG to the Athlete’s management company, “the estimate provided will be a rough estimate only”. A rough quantity estimate is in the Panel’s view, by definition an unreliable quantity estimate.

Conclusion

83. The Panel finds itself faced with compelling factual evidence and, at best, conflicting scientific evidence that acts as a double-edge sword in determining the truth. Put simply, in its assessment, the scientific evidence fails to take this storyline below the requisite of the CAS 2009/A/1926 & 1930 threshold. Therefore, the Panel reverts to the non-expert evidence and finds itself sufficiently satisfied that it is more likely than not that the presence of probenecid in the Athlete’s system resulted from kissing his girlfriend Ms. Salazar shortly after she had ingested a medication containing probenecid.

84. In consideration of the foregoing, and in contemplation of the evidence put before the Panel in both written and oral form, the Panel concludes that the Athlete has established the origin of the prohibited substance on a balance of probabilities. Furthermore, the Panel finds that even with the exercise of the utmost caution, the Athlete could never have envisioned that kissing his girlfriend of three years would lead to an adverse analytical finding for trace amounts of a banned substance that he was not familiar with. The Panel finds, therefore, that the Athlete acted without fault or negligence.

85. The appeal must therefore be dismissed.

ON THESE GROUNDS

The Court of Arbitration for Sport rules that:

1. The appeal filed by the World Anti-Doping Agency against Mr. Gil Roberts on 24 August 2017 is dismissed.

2. The decision rendered by the American Arbitration Association on 10 July 2017 is upheld.

(…)

5. All other motions or prayers for relief are dismissed.